Decoupling the Benefits of Subword Tokenization for Language Model Training via Byte-level Simulation
Paper • 2604.27263 • Published • 11
AI & ML interests
We are dedicated to advancing the field of natural language processing, in collaboration with the open-source community, through bleeding-edge research and a commitment to symbiotic development.
Recent Activity
View all activity
Papers
Decoupling the Benefits of Subword Tokenization for Language Model Training via Byte-level Simulation
Targeted Neuron Modulation via Contrastive Pair Search
gigant
authored a
paper 6 days ago
emozilla
authored a
paper 6 days ago
gigant
submitted a
paper to Daily Papers 7 days ago
samherring99
authored a
paper 8 days ago
emozilla
submitted a
paper to Daily Papers 9 days ago
samherring99
updated a
Space 9 days ago
emozilla
authored a
paper 10 days ago
bloc97
submitted a
paper to Daily Papers 13 days ago
emozilla
authored a
paper 14 days ago
bloc97
authored a
paper 14 days ago
gigant
authored a
paper 14 days ago
bloc97
submitted a
paper to Daily Papers 15 days ago
samherring99
published a
Space 17 days ago
apyh
updated a
model about 2 months ago
Rename model-00001-of-00004.safetensors to model-00069-of-00004.safetensors
2
#1 opened 3 months ago
by 
morongosteve
teknium
published a
model 3 months ago
teknium
updated a
dataset 3 months ago
teknium
published a
dataset 3 months ago
dmayhem93
authored a
paper 8 months ago
teknium
authored a
paper 9 months ago